Machine learning-based energy consumption clustering and forecasting for mixed-use buildings

© 2020 John Wiley & Sons Ltd Mixed-use buildings contribute to the sustainable development of cities by providing economic, environmental, and social benefits. Energy management of these buildings still remains a challenge due to their unpredictable energy consumption characteristics and the lac...

Full description

Saved in:
Bibliographic Details
Main Authors: Culaba, Alvin B., Del Rosario, Aaron Jules R., Ubando, Aristotle T., Chang, Jo Shu
Format: text
Published: Animo Repository 2020
Online Access:https://animorepository.dlsu.edu.ph/faculty_research/1107
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: De La Salle University
id oai:animorepository.dlsu.edu.ph:faculty_research-2106
record_format eprints
spelling oai:animorepository.dlsu.edu.ph:faculty_research-21062020-10-24T01:48:02Z Machine learning-based energy consumption clustering and forecasting for mixed-use buildings Culaba, Alvin B. Del Rosario, Aaron Jules R. Ubando, Aristotle T. Chang, Jo Shu © 2020 John Wiley & Sons Ltd Mixed-use buildings contribute to the sustainable development of cities by providing economic, environmental, and social benefits. Energy management of these buildings still remains a challenge due to their unpredictable energy consumption characteristics and the lack of design guidelines for energy efficiency and sustainability solutions. Energy consumption forecasting models have been crucial to the improvement of energy efficiency and sustainability of buildings but its application to mixed-use buildings are limited. Hence, this study aims to develop a prediction model to characterize and forecast the energy consumption of mixed-use buildings. Machine learning techniques are employed in the proposed prediction model which used k-means algorithm for clustering and support vector machines for forecasting. The prediction model was developed and demonstrated on simulated energy consumption of 30 mixed-use buildings from the open energy information database. The clustering results have found major differences in the consumption behavior of building clusters, especially on peaking characteristics. The differences were highlighted in terms of the domain knowledge on residential and commercial energy consumption behavior. The forecasting model results showed that the proposed integration of the clustering model was able to capture unique variations in the energy consumption of mixed-use buildings. This led to a 46% decrease in the mean bias error and a 10% decrease in the coefficient of variation root mean square error wherein both indicators are commonly used in building energy modeling standards. 2020-01-01T08:00:00Z text https://animorepository.dlsu.edu.ph/faculty_research/1107 Faculty Research Work Animo Repository
institution De La Salle University
building De La Salle University Library
continent Asia
country Philippines
Philippines
content_provider De La Salle University Library
collection DLSU Institutional Repository
description © 2020 John Wiley & Sons Ltd Mixed-use buildings contribute to the sustainable development of cities by providing economic, environmental, and social benefits. Energy management of these buildings still remains a challenge due to their unpredictable energy consumption characteristics and the lack of design guidelines for energy efficiency and sustainability solutions. Energy consumption forecasting models have been crucial to the improvement of energy efficiency and sustainability of buildings but its application to mixed-use buildings are limited. Hence, this study aims to develop a prediction model to characterize and forecast the energy consumption of mixed-use buildings. Machine learning techniques are employed in the proposed prediction model which used k-means algorithm for clustering and support vector machines for forecasting. The prediction model was developed and demonstrated on simulated energy consumption of 30 mixed-use buildings from the open energy information database. The clustering results have found major differences in the consumption behavior of building clusters, especially on peaking characteristics. The differences were highlighted in terms of the domain knowledge on residential and commercial energy consumption behavior. The forecasting model results showed that the proposed integration of the clustering model was able to capture unique variations in the energy consumption of mixed-use buildings. This led to a 46% decrease in the mean bias error and a 10% decrease in the coefficient of variation root mean square error wherein both indicators are commonly used in building energy modeling standards.
format text
author Culaba, Alvin B.
Del Rosario, Aaron Jules R.
Ubando, Aristotle T.
Chang, Jo Shu
spellingShingle Culaba, Alvin B.
Del Rosario, Aaron Jules R.
Ubando, Aristotle T.
Chang, Jo Shu
Machine learning-based energy consumption clustering and forecasting for mixed-use buildings
author_facet Culaba, Alvin B.
Del Rosario, Aaron Jules R.
Ubando, Aristotle T.
Chang, Jo Shu
author_sort Culaba, Alvin B.
title Machine learning-based energy consumption clustering and forecasting for mixed-use buildings
title_short Machine learning-based energy consumption clustering and forecasting for mixed-use buildings
title_full Machine learning-based energy consumption clustering and forecasting for mixed-use buildings
title_fullStr Machine learning-based energy consumption clustering and forecasting for mixed-use buildings
title_full_unstemmed Machine learning-based energy consumption clustering and forecasting for mixed-use buildings
title_sort machine learning-based energy consumption clustering and forecasting for mixed-use buildings
publisher Animo Repository
publishDate 2020
url https://animorepository.dlsu.edu.ph/faculty_research/1107
_version_ 1743177740146704384